Use this endpoint when a semantic definition must be evaluated independently across dates. Request the factor projection and set date.include_live=true when the newest active slices should be included.
Response body
application/json
These are the fields you can build against. Nested names use dot notation; optional sections are called out in their descriptions.
schemastring
Response schema name.
revisioninteger
Revision of the response format.
search_typestring
semantic.
group_bystring
Grouping dimension: date.
vector_aggregatestring
How an event's similarities to several anchor vectors combine into its score: max, mean or softmax.
per_date_limitinteger
Candidates retrieved per anchor vector and date.
min_scorenumber
Score floor applied.
totalinteger
Matched events across all dates.
candidate_eventsinteger
Candidates scored across all dates.
response_event_limitinteger
Most events one response may carry; shorten the window or raise min_score when total nears it.
resolved_vectorsnumber[][] | null
The anchor vectors embedded from vector_texts, to send as vectors in later requests; null when vectors were sent.
resolved_score_vectorsnumber[][] | null
The same for score_vector_texts; null when none were sent.
eventsobject[]
Matched events, flat; join them to groups by date.
events[].datedate
Archive date of the event.
events[].event_idstring
Event identifier.
events[].titlestring
Headline, in the original language.
events[].countrystring
Country of the event's own place (its first location; see Event details). Empty when the event has no place.
events[].scorenumber
Aggregate score over the anchor vectors.
events[].vector_scoresnumber[]
Cosine similarity to each anchor vector.
events[].score_vector_scoresnumber[]
Cosine similarity to each score vector; empty without score vectors.
events[].total_coverageinteger
Source documents in the event.
events[].total_netlocsinteger
Distinct source sites in the event.
events[].coverage_concentration_indexnumber
Concentration of the coverage across sites, from 0 to 1.
events[].materiality_scorenumber
Continuous materiality from 0 to 1.
events[].sentimentstring
positive, neutral or negative.
groupsobject[]
One summary per date scanned.
groups[].datedate
The date.
groups[].matched_countinteger
Matched events on the date.
groups[].candidate_countinteger
Candidates scored on the date.
groups[].candidate_retrieval_hitsinteger
Candidates the vector index returned for the date, before deduplication across anchor vectors.
groups[].truncatedboolean
Whether the date's candidate budget filled, so more matches may exist.
groups[].eligibleobject
All events on the date that pass the filters, matched or not: count, total_coverage, total_netlocs, total_materiality and total_sentiment_weight. Use them as denominators.
date_windowobject
from, to, include_live and archive_cutoff.
statsobject
Totals over the dates: dates requested and scanned, eligible, candidate and matched events, and truncation.
timing_msobject
Server time in milliseconds.
Responses and errors
HTTP
200Request succeeded. The response body is shown in the panel on the right.
400invalid_request — The grouped request is invalid or omits required date, vector, or projection fields.
401api_key_required — No API key was sent. Send it as Authorization: Bearer <key> or the api-key header.
401invalid_api_key — The API key is invalid or has been revoked.
403access_denied — The requested archive date is outside the caller's World tier.
404not_found — The requested date, entity, ticker, or event does not exist in the accessible index.
410cursor_expired — The cursor was minted against a replaced index build. Restart pagination from the first page.
429rate_limited — The request exceeded the endpoint or account rate limit. Respect Retry-After when supplied.
501backend_not_configured — The World backend is not configured for this deployment.
503api_key_auth_unavailable — API key authentication is temporarily unavailable; retry shortly.
502backend_error — The BFF could not obtain a valid response from the World backend.
504backend_timeout — The World backend or an upstream data source timed out.
422grouped_response_too_large — The request exceeds the bounded grouped response budget; shorten the window or raise the score floor.
503grouped_search_busy — The grouped-search worker pool is full; retry after backoff.